Abstract S4-04: Overcoming Resistance to PI3K Inhibitors in PIK3CA Mutant Breast Cancer Using CDK4/6 Inhibition: Results from a Combinatorial Drug Screen

SR Vora,N Kim,C Costa,EL Lockerman,X Li,Y Chen,A Cao,M Pinzon-Ortiz,M Liu,S Kim,R Schlegel,A Huang,JA Engelman

DOI: https://doi.org/10.1158/0008-5472.sabcs13-s4-04

IF: 11.2

2013-01-01

Cancer Research

Abstract:Abstract Various components of the phosphatidylinositol 3-kinase (PI3K) pathway are deregulated across a spectrum of human cancers. Notably, cancers with PIK3CA mutations, seen in roughly 30% of breast cancers, are amongst the most sensitive to PI3K inhibitors (PI3Ki) as single agents. Therefore, there have been great efforts to develop PI3K inhibitors specifically for these types of cancers, and many agents have already entered the clinic. Although initial responses and prolonged stable disease have been observed, resistance frequently emerges. Moreover, there is a subset of PIK3CA mutated cancers that unexpectedly do not exhibit an initial response or disease stabilization upon exposure to PI3K inhibitors, despite presence of the mutation. These cancers are said to have de novo resistance to PI3K inhibition. To determine methods of overcoming resistance to PI3K inhibitors, we generated two models with acquired resistance to the p110a isoform specific inhibitor BYL-719 (BYL) using MDA-MB-453 (453) and T47D. We also established one model of resistance to the pan-PI3K inhibitor GDC-0941 using MCF7 cells. These lines were chosen because of their PIK3CA mutated status and sensitivity to PI3K inhibition. Each cell line was grown in increasing concentrations of PI3K inhibitor until the cells proliferated readily at a dose of drug that effectively reduced cell viability and inhibited pAKT in the sensitive parental cell lines. Interestingly, both BYL resistant cells (453R and T47DR) were cross resistant GDC and the MCF7R line was refractory to BYL. To elucidate mechanisms to overcome resistance to PI3K inhibitors, we undertook a combinatorial drug screen, in which PI3K inhibitor resistant cells were treated with escalating doses of a panel of 45 targeted agents, both in the presence and absence of a fixed dose of PI3Ki, to determine which agents synergized effectively with PI3K inhibition in these resistant cells. We observed in each of the three PI3Ki resistant models a synergy between the CDK4/6 inhibitor LEE-011 and PI3K inhibition. We furthermore tested this combination of agents in a PIK3CA mutated breast cancer model with de novo resistance to PI3K inhibitors, CAL51, and again noted efficacy with the combination of GDC and LEE-011 while either agent on its own displayed minimal activity. To determine whether addition of CDK 4/6 inhibition might be an effective addition to PI3Ki in the upfront setting in vivo, we injected each of the PIK3CA mutated lines MCF7, 453, and T47D into female nude mice and treated with vehicle, BYL, LEE-011, or the combination. We noted in each of the three models that the combination of agents, led to tumor regression that was more substantial than single agent treatment, and furthermore delayed the acquisition of resistance relative to single agent therapy. We furthermore tested GDC with LEE-011 singly and in combination in both MCF7 and CAL51 xenografts and again noted that the combination of agents led to tumor regression, whereas in these instances, single agent treatment did not. We conclude that the combination of PI3K and CDK 4/6 inhibition may be an effective strategy for treating PIK3CA mutated breast cancer and deserves further study in the clinical setting. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S4-04.